I work for NVIDIA and help out at CMU, JHU and Stanford in an effort to make Bioinformatics Better. Everything I post is my personal opinion. You're awesome!
Pittsburgh, PA
Joined July 2013
- Tweets7K
- Following1.4K
- Followers2.1K
- Likes7.3K
Pinned Tweet
If you are interested in federating biobank data -- come to Copenhagen in September! nordic-conference-on-future-…
3
136
Ben Busby! retweeted
"Targeted ANGPTL3 inactivation in nonhuman primates was well-tolerated and led to a > 90% mean reduction in circulating ANGPTL3 protein, with durability confirmed out to 3 years following a single dose." nature.com/articles/s41586-0…
Ben Busby! retweeted
RGC contributed association summary data to a new @Nature study led by @VUamsterdam. The study found 48 new genetic regions associated with Alzheimer’s disease, with risk concentrated in microglia and neurons.
1/n
Out today in @Nature: our Alzheimer’s disease GWAS of ~2.8M people (~184k cases). We find 48 new loci, ~3x higher SNP-heritability, enrichment in genes up-regulated in microglia and down-regulated in neurons, and we release public summary statistics. nature.com/articles/s41586-0…
Check out this splice variant model: pubmed.ncbi.nlm.nih.gov/4210… !!
Ping me or with questions!
Ben Busby! retweeted
Explore the slides from HPRC’s #ASHG2026 workshop, “The Human Pangenome Release 3: Data, Tools, Applications, and Ethics,” featuring an overview of the resources, tools, and research applications shared during the session.
View the slides: drive.google.com/file/d/1pv3…
Ben Busby! retweeted
Great thread and video summary from @eigenNick explaining his new work #EMBER ! A core idea is to learn gene representations that capture both molecular function and cellular context - then reuse them compositionally to predict perturbation effects, genetic interactions, and drug responses 🚀 biorxiv.org/content/10.64898…
Contextualized gene representations work surprisingly well for predicting multigene perturbation effects.
I’m excited to introduce EMBER, a pretrained gene-centric foundation model designed specifically to learn context-specific gene representations. A joint work with @jmuiuc and @ericxing.
With innovations in single-cell pretraining and dataset curation, EMBER gene representations can be combined with a lightweight probe for single gene knockouts, genetic interactions (double knockouts), and even drug perturbations.
Both a 🧵 and a 🎥, got in front of a camera to share some highlights! (1/n)
Ben Busby! retweeted
A fantastic turnout for HPRC's workshop at #ASHG2026!
Thank you to everyone who joined us to learn more about the Human Pangenome Release 3 and its data, tools, applications, and path forward.
Ben Busby! retweeted
In my role as PhD advisor, I'm nowadays just telling my students to pick an important problem:
1. That is not trivially solvable by AI today
2. That is not theoretically unsolvable
3. For which a solution technique will just get better with AI advancements, not obsolete.
Ben Busby! retweeted
HERE is the thread you're actually looking for!
Avengers assemble: 1 year BakerLab + NVIDIA GenAIR = 🔥 rfd4-proteina.org
Presenting RFD4-Proteina, our latest protein design model, significantly better and faster than everything we compared against, both in silico and in the lab! 1/30
Ben Busby! retweeted
Introducing RFD4-Proteina! A monumental advance in protein design! Diverse frontier capabilities across design tasks and biological modalities from the combined strength of the @UWproteindesign RFdiffusion and @NVIDIAHealth Proteina lineage of models. Available open source, through our partners, and as an NVIDIA NIM.
NVIDIA and @UWproteindesign introduce RFD4-Proteina, an open model to design proteins and refine candidates for lab testing, paving the way for new medicines and diagnostics.
Announced at GTC Berlin: nvda.ws/4ywvAss
Ben Busby! retweeted
A prevention conundrum:
- LDL-C lowering is prioritized after an ASCVD event because near-term absolute risk is high.
- Yet relative benefit is greater before a first event and at younger ages.
sciencedirect.com/science/ar… (Burger PM et al Atherosclerosis 2024)
We need to keep moving toward improving high lifetime-risk identification before the disease declares itself.
Ben Busby! retweeted
Using low-coverage DNA sequencing and potentially expanding screening options, researchers developed a method to detect Robertsonian translocations, chromosome changes linked to fertility challenges and increased cancer risk.
Read the preprint: biorxiv.org/content/10.64898…
Ben Busby! retweeted
Excited to discuss the translational impact of global genetics during the #ASHG26 Distinguished Speakers Symposium "Globalizing Human Genetics: From Discovery Through Implementation" Oct 24 with @EimearEKenny @gabecasis @genandgenes M. Ramsay @OOjewunmi - Hope to see you there!
@Amgen
Ben Busby! retweeted
AI agents are moving from engineering workflows to the fab floor.
Announced at #SEMICONWest, NVIDIA cuPotential joins the CUDA-X portfolio, helping developers build AI agents that simulate atomic and molecular interactions so engineers can model semiconductor materials and manufacturing processes faster.
Ben Busby! retweeted
Building on decades of joint work with national labs, NVIDIA has committed $1B towards developing advancements in US science. For research across quantum computing, healthcare and energy.
Many years ago, as a postdoc at lawrence berkeley I saw how industry partners were integral to applied research at the federal level and also critical to training the scientists and engineers of the future
Ben Busby! retweeted
Molecules in the morning, robots in the afternoon. Two hands-on #NVIDIAGTC Berlin labs on October 22.
🧬 9:00 - Drug discovery with BioNeMo nvda.ws/4AZMi5O
🥨 Lunch - currywurst break
🤖 13:00 - Medical robotics with Isaac for Healthcare nvda.ws/4AY0fBd
Ben Busby! retweeted
Metrics of OpenEvidence use are incredible - 42 million times in August
Clinicians (me included) used @OpenEvidence 42 million times in August, up from 18 million in December.
A couple thoughts:
1 - evaluation — When I wrote about it in March, the research behind it was already too thin to say whether it improves clinical decisions or just confirms the ones doctors already made. Usage has more than doubled since then, and I can't find any new independent evaluation.
2 - global health — @Anthropic and OpenEvidence will give a free, locally adapted version to clinicians in about 100 lower-income countries, starting with Uganda, Angola, Sudan, Haiti, and Mongolia.
A doctor in rural Sudan getting the same answer engine as an oncologist at MSK (or me, a pediatrician in DC) could be a very good thing. But I'd still like to see more evaluation.
buff.ly/XKM2Qjl
#DrYou #HealthcareAI #GlobalHealth
Ben Busby! retweeted
Have you ever tried to reuse someone else's spatial-omics dataset and given up halfway? 🔄
A team from Helmholtz Munich and Lamin is running a short community survey on this (spatial transcriptomics, multiplexed imaging and related methods), to help offer potential solutions on what could make it easier. Whether you generate the data, analyse it or manage it, we'd like to hear from you. PhD students, postdocs, PIs, core facility staff and industry are all welcome.
→ 5–10 minutes
→ Anonymous: no name, email or IP address collected
→ Anonymised results may be published openly
Your answers will shape an open, cloud-native repository for spatial-omics data and inform a research funding proposal. Take the survey: docs.google.com/forms/d/1aZW…
📢 Do share with anyone in your lab or network who works with spatial data!
Ben Busby! retweeted
Biotia Raises New Oversubscribed $9M Financing and Expands Board to Scale Infectious Disease Diagnostics in Women's Health and Orthopedics prnewswire.com/news-releases… Very excited to expand our diagnostic tools and patient impact at @BiotiaInc !
Ben Busby! retweeted
A video summary of our recent preprint
biorxiv.org/content/10.64898…
Ben Busby! retweeted
Most drugs work like a key in a lock. But hard-to-treat diseases often lack a lock to begin with, leaving them out of reach for conventional medicine.
Covalent drugs change the game by bonding permanently to their targets. To find the right molecules, @ExpeditionMedicines fine-tuned the AIMNet2 model with the NVIDIA ALCHEMI Toolkit—screening 15 million structures per $1,000 of compute (17x faster and cheaper than GFN2-xTB).
expeditionmedicines.com/news…
